Deep Convolutional neural network (CNN) in tea leaf chlorophyll estimation: A new direction of modern tea farming in Assam, India

This study presents the uprising of leaf chlorophyll estimation from traditional mechanical method to machine learning-based method. Earlier chlorophyll estimation techniques such as Spectrophotometer and Soil Plant Analysis Development (SPAD) meter demand cost, time, labour, skill, and expertise. A small-scale tea farmer may not afford these devices. The present study reports a low-cost digital method to predict the tea leaf chlorophyll using 1-D Convolutional Neural Network (1-D CNN). After capturing the tea leaf images using a digital camera in a natural light condition, a total of 12 different colour features were extracted from tea leaf images. A SPAD was used to estimate the original chlorophyll value of the tea leaves. The paper shows the correlation of original tea leaf chlorophyll with the extracted colour features of the tea leaf images. Apart from 1-D CNN, the Multiple Linear Regression (MLR) and K-Nearest Neighbor (KNN) were also applied to predict the tea leaf chlorophyll and compared their results with the 1-D CNN. The 1-D CNN model outperformed with an accuracy of 81.1%, Mean Absolute Error (MAE) of 3.01, and Root Mean Square Error (RMSE) of 4.18. The investigation system is very simple and cost-effective. It can be used in tea farming as a digital SPAD for faster and accurate leaf chlorophyll estimation in an easy way.

Leaf Diseases Classification on Peanut Leaves Based on Texture and Colour Features

Peanuts are a food crop commodity that Indonesians widely consume as a vegetable fat and protein source. However, the quality and quantity of peanut productivity may decline, one of which is due to plant diseases. Efforts that can be made to maintain peanut productivity are the application of technology to detect peanut plant diseases early; thus, disease control can be carried out earlier. This study presents a technology development application, particularly digital image processing, to identify disease features of infected peanut leaves based on GLCM texture features and colour features in the HSV colour space and classified using the SVM method. The development of the SVM method that is applied is the Multiclass SVM with the DAGSVM strategy, which can classify more than two classes. Based on the experimental results, it confirms that the combination of HSV colour features and GLCM texture features with an angular orientation of 0 degrees and classified by the Multiclass SVM method with polynomial kernels produces the highest accuracy, i.e. 99.1667% for leaf spot class, 97.5% for leaf rust class, 98.8333% for eyespot class, 100% for normal leaf class and 100% for other leaf class.

A Novel Frame Work for Detection of Chemically Ripened Mango Fruits Using Dominant Colour Descriptors

Nowadays computer vision systems are widely used for identification, classification and grading of different kind of fruits. Existing research concentrates on features like size of the fruit, colour, shape and texture for classification and maturity detection of mango fruits. Colour of the fruit is one of the prominent feature, though a lot of effort is focused towards identification, maturity and defect detection of mango using colour features,less attempts are made in the direction of the identification of artificially ripening of mango fruits.From another perspective very few attempts are concentrated towards deeper analysis of colour features of the fruits. In order to solve these issues this research paper proposes a new framework for detection of artificial ripening of mango fruit based on MPEG-7 colour descriptors. The proposed scheme includes three stages: first, a pre-processing stage consisting of masking, filtering, segmenting and cropping of an image followed by dominant colour extraction using dominant colour descriptors which are finally mapped with the help of clustering to identify the artificial ripening of mango fruit. The results of experimentsis carried out on two different datasets involving four types of mangoes which demonstrates robustness and efficiency of the proposed method against various other methods.